Cows and IoT: A moo-t point?

If you've read my previous articles, you'll see that I've taken some interest in researching Teltonika's TAT100 (or just the TAT, as I refer to it). Seeing what the TAT can do, and the kind of avenue it has opened up for tracking non-powered assets on a low/no-installation basis is absolutely revolutionary.

One of my colleagues, particularly jazzed by this device (I mean, it's a battery-powered asset tracker for goodness sake), and I started brainstorming all the things that we could strap the TAT to - your husband, your dog, fling it into your teenager's bag before they go for a "sleepover at Sarah's house", or strap it to a cow.

The latter is something I'm particularly interested in. Agriculture is one of the biggest growing sectors in South Africa - in fact, for the first time since 2001, it is booming. IoT is catching on and farmers everywhere are starting to get their hands on the greatest advantages that tech can provide (and it's pretty cool). I mean, I'm not a typical tractor-loving person, but even I can appreciate this big beauty from Jon Deere. Who could imagine that technologies are already affording farmers the opportunity to control their tractors from their smartphones?

Although 4iR technologies are increasing exponentially, the application of IoT within the agricultural industry still has a long way to go within South Africa. There is a lot of talk about NB-IoT and how that could change the game for wide-area networking for small package data transmissions, like location-based tracking. Something we've been researching is the potential to help other businesses track their assets periodically - as in not real-time tracking (which is often very wasteful of battery-powered trackers and data-intensive). This might be something like a location-based check-in or a predetermined report to be delivered every 3-6 hours or so (this criterion differs depending on who is asking, and for what kind of Agri case).

We've been interviewing business owners from different sectors to discuss the proverbial thorns in their sides. One conversation we had recently (and this is where agritech comes in) is related to tracking medium to large assets - namely cows.

A colleague and I were bouncing ideas back and forth to predict the best way to wirelessly track non-powered, but widely-distributed assets (in this case, cows) in the middle of nowhere. No doubt you would be using GSM based infrastructure - especially in South Africa. But the part we got stuck on is the hardware element... and here's why:

Some Assumptions I Made About Tracking Livestock

"Just use a collar"

That was my first suggestion. Take the battery-powered tracker, program it to transmit its location every 6 to 8 hours and strap it around the cow's neck. Now, you have all your cows accounted for. But no...

Have you ever tried to collar a cow? But have you ever tried to collar 300 cows? I'm not going to assume your personal experiences but I certainly have not. But the multiple farmers that I spoke to on the phone have. And let me tell you: the experience is not unlike trying to get your cat to swallow its antibiotics. You're also not collaring one cow alone - you're collaring at least 100. And these trackers are battery powered - which means, at some point (depending on the number of reports you request, the sun exposure, et al.) you will have to recall those cows and either:

A) Recharge your trackers (for the ones with rechargeable batteries) or,

B) Replace the tracker altogether/change out the disposable batteries.

The saying goes that there is no rainbow without a little rain, so maybe the collaring and re-collaring is just something that farmers need to suck up to get what they want - angry farmers do suggest otherwise.

"Put tags on the cows' ear's"

This was my second suggestion. Using RFID tags on cows and setting up base stations as check-in points is not a new invention. It's been done a couple of times, and I had the opportunity to chat with some people that have experience with this specific use case - which is how I learnt about how this is not the best solution depending on your requirements. Although RFID does solve the "recall and recollar" issue (tags supposedly last up to 30 years), there are other factors that make this solution somewhat unreliable.

The first is that you have to set up RFID base stations. This requires a fair amount of infrastructure setup and a lot of strategic placement as to where exactly you want to place these base stations. The base stations need access to power - so if you're looking at 18 000 hectares of farmland, you may need to consider a scalable power source.

Another factor to take into consideration is that in order to be identified by the base station, cows have to stand roughly within a 30-meter radius of the antenna. You'd think that cows would manage this distance, but farmers suggest that getting cows close enough to the antenna is a bit of a hit and a miss when it comes to getting reliable information.

Lastly: Have you ever had an itchy ear, or have you seen your dog scratch its ear? It's quite aggressive. Cows do that too. They also rub their head against trees, each other, or their hooves, which sometimes damages the ear tag.

Maybe RFID could work, if your circumstances were juuuust right (like with Goldilocks and the three bears).

Conclusion(s)

What it came down to, is that every farmer I spoke with, wanted the following thing: they wanted a low-cost tracker, with low/no installation, that is able to send data every 3 hours. Oh, and they would like it battery-powered, please. Now, unless Teltonika's revolutionary thimble-size crumb of a tracker actually exists (spoiler: it doesn't), then we may have to do a little more R&D into this situation.

I suppose if you've gotten this far down the post that you are expecting me to make a revolutionary recommendation as to what I think the solution is to track cattle - I don't. But if I find out, I will be sure to share it.

*** AN EXTRA BIT ABOUT ME***

Only discovering the world of IoT and hardware roughly a year ago comes with a fair number of challenges: mostly related to the fact that I have next to no experience in the field, and am constantly Googling my way through every use-case scenario and setup when it comes to IoT solutions & hardware. However, what I lack in experience, I make up for in honesty, curiosity, and the absorbency of my sponge-like brain. Everyone has to start somewhere, and the reality is that the majority of the businesses needing, and undeniably benefitting, from IoT or other nifty technologies, didn't even know they needed it, to begin with. I'm here to make IoT more accessible for newbies, make technical jargon less immobilising, shed some light on some interesting use cases, and learn about how businesses can leverage their assets & data in the universe of end-to-end IoT.